Andrew Selbst on Justifying Algorithmic Decisionmaking
In this episode, Andrew Selbst, a Postdoctoral Scholar at Data & Society Research Institute and Visiting Fellow at the Yale Information Society Project, discusses his article "The Intuitive Appeal of Explainable Machines" (co-authored with Solon Barocas, Assistant Professor in the Department of Information Science at Cornell University), which will appear in the Fordham Law Review. Selbst begins by framing the promise and peril of algorithmic decisionmaking. Among other things, he explains how algorithmic decisionmaking works and describes the current debate over how to regulate it. In particular, he notes that many regulatory proposals focus on requiring the explanations of how an algorithm works. But he and Barocas argue that regulators should also require justifications for the construction of those algorithms, and propose some ways in which those justifications could be provided.
Keywords: algorithmic accountability, explanations, law and technology, machine learning, big data, privacy, discrimination